This question is prompted by two very similar examples reported online: this one (also on Twitter) and this one. In both, teachers introduce aspects of the political system described in the 1949 novel Nineteen Eighty-Four to a classroom in which they are (or will shortly be) studying the novel, as an aid to understanding the issues it raises. Essentially, the teachers ask students to secretly or publicly report on the behaviour of other students, introduce points systems and punishments, etc.

In both cases, as far as I can tell, the students are unaware that the experiment is going on, and it's more of an exercise than an experiment; its purpose is not to gather any data or test a hypothesis.

My question is: from a legal and an ethical review standpoint, what would a lecturer typically need to do before introducing these sorts of activities into their class, in the way described above, as part of a learning exercise?

I am not asking for personal advice here, I'm just interested in how this sort of thing would be assessed by a typical ethical review panel. I work at a research institute where we do have a very well-defined ethical review process but only for animal studies; we don't do research involving human subjects. Furthermore, several of the steps we would normally be expected to provide (sample size, etc) do not actually apply here since the purpose is not actually research, it is teaching "pretending" to be research: the experiment does not need to generate any meaningful data to fulfil its purpose (unless you redesigned it as a study of the effectiveness of this method in raising understanding of certain issues).

Several of the comments on Twitter noted that this would probably be considered unethical as described; the participants have not consented to be involved, for example. Does the fact that the 'experiment' is not (scientifically) actually an experiment make any difference? Lastly, these experiments sound like they are high-school environments; does the answer change depending on whether the students concerned are above or below 18?

Somewhat related: Ethics of conducting research on a class.

  • I'm looking for answers that identify what the key factors would be in this general situation, so this question is intentionally light on specifics; if you think I need to add more detail to the hypothetical situation to make this answerable, please be specific about what those details should be.
    – arboviral
    Feb 20, 2018 at 13:20
  • 1
    This sounds like a famous novel: en.wikipedia.org/wiki/The_Wave_(novel) Feb 20, 2018 at 14:04
  • 2
    I'm not in a field that uses human subjects, but my impression has been that IRBs regulate research only (loosely defined as "that which will eventually be published in a research journal/conference") and have no role at all in overseeing teaching per se. There's no formal ethics pre-review for teaching that I've ever heard of, Feb 20, 2018 at 14:23
  • 1
    There is, of course, a difference between performing an experiment on students and teaching them a lesson. Taken to an extreme, you could also be questioning a teacher trying out a new curriculum technique - there the students actually are part of an experiment...
    – Jon Custer
    Feb 20, 2018 at 14:24
  • 1
    @henning actually I initially assumed it was a reference to the Stanford Prison Experiment, but I have now discovered it's basically identical to the (real) 1967 exercise called The Third Wave, which resulted in the film The Wave, which resulted in the novel The Wave. So it's based on the experiment that the film the novel was based on, was based on. :)
    – arboviral
    Feb 20, 2018 at 14:50

2 Answers 2


Virtually every modern university has an ethics procedure for experiments conducted on human subjects. If an academic wants to conduct an experiment on students (especially one that involves deception) they would generally need to put in a proposal to their university ethics board. This would disclose the details of the proposed experiment, and the nature of any deception of subjects, and the university would make a decision as to whether or not to allow the experiment to proceed.

Experiments involving deception of subjects are often approved (many psychological experiments are of this kind), but it is usual to require that the subjects be "debriefed" after the experiment, to disclose to them the deception they were subjected to, and how the experiment worked. In cases where the deception is likely to be distressing to the subjects, or the gains from the experiment are minor, the ethics board might decide not to allow the experiment to proceed, or might require it to be altered in some way.

Now, with regard to the particular situation described in the Atlantic article, this does not appear to be an academic experiment at all; it is an extended High School class exercise that uses some mild deception to illustrate the world of the book through mimicking aspects of it in the conduct of the classroom. Students are subjected to some overbearing actions by the teacher to give them a small taste of totalitarianism. Nothing described in the article would be unlawful, since at worst it involves cases of nasty looks, overbearing control, or dressing down a dissenter.

High Schools generally do not have formal ethics panels like the universities, since they do not conduct academic research. It is likely that there is significant teacher discretion on how to conduct classes, but this would be subject to oversight by the school principal. In this case I think it is unlikely that the exercise would have been subjected to any preliminary scrutiny of the kind that would occur in a university ethics application. Obviously students would be free to object to this teaching exercise and take the matter up with the school principal, but this does not appear to be a case of experimentation that would go to an ethics board.

The basic ethical question at play here is: is it okay to be overbearing and mildly bullying to your students in order to give them a genuine taste of what totalitarianism is like. Personally, I see no problem with the exercise, since the likely distress to students is mild in comparison to the value of the learning experience, and the students are made aware of the deception after the exercise is completed, so that they can discuss it in relation to the book. Others may disagree, but to me it sounds like a very creative and useful teaching exercise, and the teacher doesn't take it too far. I think it is quite brilliant, and I would happily enrol my own children in this class!

  • This is a good answer as far as it goes (+1), but it would be useful to highlight whether there are any activities that might reasonably form part of this sort of exercise that would in fact be illegal or result in disciplinary action (e.g. discrimination against a minority, collection of personal data, etc). I can easily see someone deciding to try this in their classroom but modifying it slightly, and I wonder how easy it would be to do so in a way that gets you in trouble.
    – arboviral
    Feb 21, 2018 at 8:33
  • 2
    I think it is unlikely that discrimination per se would be unlawful in this context, so long as the discrimination is used as a teaching tool to allow the participants to experience and see the discrimination that occurs under the system that is being taught. The relevant question would be: does the occurrence of discrimination in the exercise constitute discrimination in regard to the quality of educational service being provided to students, or does it merely assist all students in understanding the nature of the system being taught.
    – Ben
    May 1, 2018 at 6:08

Short answer: You have to make sure no one cries. If you create an environment that results or reasonably could be expected to result in emotional hurt, then you've been unethical.

At university, there will be a "human subjects" review of some sort. Anyone writing a grant or proposal for anything will have to be passed by this review. I've never done experiments on human subjects, but every grant I've received required an additional piece of paper which stated that there were no human subjects involved (in my math research.)

In your case, as you say, it's not really an experiment, but a teaching technique. I don't know what to say about this, exactly, but I would not be surprised to find that the human subjects panel or office or whatever also oversees such things. No doubt you have to be careful.

Story from 1970: I was in 5th grade in a small town in Nebraska which had a teachers college. Every master's of ed candidate at the college had to do something for a thesis and often this was some sort of experiment on a classroom full of students. Lucky us; those of us in town were the handiest subjects for all these theses. As I moved through the Kearney, Nebraska school system, I was the subject of 4 or 5 experiments each year. We were video-taped, quizzed, surveyed, lied to, etc., on a regular basis.

In 5th grade, my teachers along with some master's candidate thought it would be instructive if they did the "Blue-eyed /Brown-eyed" thing to us. They announced one morning that a newspaper article stated, scientifically, that brown-eyed children were smarter than blue-eyed children. Then then segregated us and the smart kids, not needing so much instruction, got to play games and have snacks. While the inferior kids had to do extra math work sheets, you know, because they needed the extra practice, being so dumb and all.

Of course this was supposed to teach the (100% white) children about racial discrimination and how bad it hurt one's feelings. We had girls in tears by the end of the day. The next day, there were some red-faced daddies with their fists cocked at the school. "You make my daughter cry; I break your ribs." I tell this story to illustrate the ethical problem.

I don't think the situation would be much different with college freshlings. If you hurt their feelings too much, Daddy is likely to visit you. Set aside whether the human subjects panel approved of your experiment or whether there are legal remedies.

  • "I would not be surprised to find that the human subjects panel or office or whatever also oversees such things." Hm. I would be very surprised. Feb 20, 2018 at 15:41
  • 8
    Well, for instance, I've been involved in some educational research. If, as part of an educational study, I want to give my class a calculus quiz and write a paper about how they did, that requires IRB approval, informed consent from the students, and a way for them to opt out. If I want to give them that very same quiz to assess their learning, for the purposes of evaluating my own teaching, no approval or consent is required, and I can make it mandatory. Feb 20, 2018 at 15:52
  • 4
    @B.Goddard I'm sorry for misinterpreting what you said. Whether or not you are willing to credit it, there are many things that go into how people understand and interpret statements besides the words used. Feb 20, 2018 at 20:18
  • 2
    I would also point out that the "nobody cries" standard is not necessarily upheld by traditional teaching either - you should see our department after a calculus exam :-) Feb 21, 2018 at 4:18
  • 1
    @NateEldredge "Nobody cries" was not put forth as a standard for anything except judging your experiments. I keep an empty box of Kleenex on my desk, because I'm so heartless. If someone cries because they bombed your exam, then it was a good exam. If someone cries because you tricked them, then you've probably been unethical.
    – B. Goddard
    Feb 21, 2018 at 13:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .